February 2023 - Wine-commits

Zebediah Figura : vkd3d-shader/hlsl: Parse the tex2D() intrinsic.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: 645ec0167aebf529408a60437a7f5bcd724cfd6d URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/645ec0167aebf529408a60437a7f5… Author: Zebediah Figura <zfigura(a)codeweavers.com> Date: Wed Aug 11 18:26:21 2021 -0500 vkd3d-shader/hlsl: Parse the tex2D() intrinsic. --- libs/vkd3d-shader/hlsl.y | 61 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 61 insertions(+) diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index e0e0e5ce..4731d519 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -3067,6 +3067,66 @@ static bool intrinsic_step(struct hlsl_ctx *ctx, return !!add_implicit_conversion(ctx, params->instrs, ge, type, loc); } +static bool intrinsic_tex(struct hlsl_ctx *ctx, const struct parse_initializer *params, + const struct vkd3d_shader_location *loc, const char *name, enum hlsl_sampler_dim dim) +{ + struct hlsl_resource_load_params load_params = {.type = HLSL_RESOURCE_SAMPLE}; + const struct hlsl_type *sampler_type; + struct hlsl_ir_resource_load *load; + struct hlsl_ir_load *sampler_load; + struct hlsl_ir_node *coords; + + if (params->args_count != 2 && params->args_count != 4) + { + hlsl_error(ctx, loc, VKD3D_SHADER_ERROR_HLSL_WRONG_PARAMETER_COUNT, + "Wrong number of arguments to function '%s': expected 2 or 4, but got %u.", name, params->args_count); + return false; + } + + if (params->args_count == 4) + { + hlsl_fixme(ctx, loc, "Samples with gradients are not implemented.\n"); + } + + sampler_type = params->args[0]->data_type; + if (sampler_type->type != HLSL_CLASS_OBJECT || sampler_type->base_type != HLSL_TYPE_SAMPLER + || (sampler_type->sampler_dim != dim && sampler_type->sampler_dim != HLSL_SAMPLER_DIM_GENERIC)) + { + struct vkd3d_string_buffer *string; + + if ((string = hlsl_type_to_string(ctx, sampler_type))) + hlsl_error(ctx, loc, VKD3D_SHADER_ERROR_HLSL_INVALID_TYPE, + "Wrong type for argument 1 of '%s': expected 'sampler' or '%s', but got '%s'.", + name, ctx->builtin_types.sampler[dim]->name, string->buffer); + hlsl_release_string_buffer(ctx, string); + } + else + { + /* Only HLSL_IR_LOAD can return an object. */ + sampler_load = hlsl_ir_load(params->args[0]); + + load_params.resource = sampler_load->src; + } + + if (!(coords = add_implicit_conversion(ctx, params->instrs, params->args[1], + hlsl_get_vector_type(ctx, HLSL_TYPE_FLOAT, hlsl_sampler_dim_count(dim)), loc))) + coords = params->args[1]; + + load_params.coords = coords; + load_params.format = hlsl_get_vector_type(ctx, HLSL_TYPE_FLOAT, 4); + + if (!(load = hlsl_new_resource_load(ctx, &load_params, loc))) + return false; + list_add_tail(params->instrs, &load->node.entry); + return true; +} + +static bool intrinsic_tex2D(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc) +{ + return intrinsic_tex(ctx, params, loc, "tex2D", HLSL_SAMPLER_DIM_2D); +} + static bool intrinsic_transpose(struct hlsl_ctx *ctx, const struct parse_initializer *params, const struct vkd3d_shader_location *loc) { @@ -3163,6 +3223,7 @@ intrinsic_functions[] = {"smoothstep", 3, true, intrinsic_smoothstep}, {"sqrt", 1, true, intrinsic_sqrt}, {"step", 2, true, intrinsic_step}, + {"tex2D", -1, false, intrinsic_tex2D}, {"transpose", 1, true, intrinsic_transpose}, };

1 0

Zebediah Figura : vkd3d-shader/hlsl: Parse the tex3D() intrinsic.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: 7c1c2e6cc4f9d2f6e8df0a46a58c721a74ba5dff URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/7c1c2e6cc4f9d2f6e8df0a46a58c7… Author: Zebediah Figura <zfigura(a)codeweavers.com> Date: Wed Aug 11 18:27:57 2021 -0500 vkd3d-shader/hlsl: Parse the tex3D() intrinsic. --- libs/vkd3d-shader/hlsl.y | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/libs/vkd3d-shader/hlsl.y b/libs/vkd3d-shader/hlsl.y index 4731d519..6c618912 100644 --- a/libs/vkd3d-shader/hlsl.y +++ b/libs/vkd3d-shader/hlsl.y @@ -3127,6 +3127,12 @@ static bool intrinsic_tex2D(struct hlsl_ctx *ctx, return intrinsic_tex(ctx, params, loc, "tex2D", HLSL_SAMPLER_DIM_2D); } +static bool intrinsic_tex3D(struct hlsl_ctx *ctx, + const struct parse_initializer *params, const struct vkd3d_shader_location *loc) +{ + return intrinsic_tex(ctx, params, loc, "tex3D", HLSL_SAMPLER_DIM_3D); +} + static bool intrinsic_transpose(struct hlsl_ctx *ctx, const struct parse_initializer *params, const struct vkd3d_shader_location *loc) { @@ -3224,6 +3230,7 @@ intrinsic_functions[] = {"sqrt", 1, true, intrinsic_sqrt}, {"step", 2, true, intrinsic_step}, {"tex2D", -1, false, intrinsic_tex2D}, + {"tex3D", -1, false, intrinsic_tex3D}, {"transpose", 1, true, intrinsic_transpose}, };

1 0

Francisco Casas : vkd3d-shader/hlsl: Leave offset empty for array and struct derefs.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: 315966dc2186b15e5bbe794d7d0610e4c809212e URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/315966dc2186b15e5bbe794d7d061… Author: Francisco Casas <fcasas(a)codeweavers.com> Date: Thu Feb 9 20:29:17 2023 -0300 vkd3d-shader/hlsl: Leave offset empty for array and struct derefs. --- libs/vkd3d-shader/hlsl_codegen.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/libs/vkd3d-shader/hlsl_codegen.c b/libs/vkd3d-shader/hlsl_codegen.c index c6d43928..46444da5 100644 --- a/libs/vkd3d-shader/hlsl_codegen.c +++ b/libs/vkd3d-shader/hlsl_codegen.c @@ -125,6 +125,7 @@ static struct hlsl_ir_node *new_offset_instr_from_deref(struct hlsl_ctx *ctx, st static void replace_deref_path_with_offset(struct hlsl_ctx *ctx, struct hlsl_deref *deref, struct hlsl_ir_node *instr) { + const struct hlsl_type *type; struct hlsl_ir_node *offset; struct hlsl_block block; @@ -134,6 +135,16 @@ static void replace_deref_path_with_offset(struct hlsl_ctx *ctx, struct hlsl_der /* register offsets shouldn't be used before this point is reached. */ assert(!deref->offset.node); + type = hlsl_deref_get_type(ctx, deref); + + /* Instructions that directly refer to structs or arrays (instead of single-register components) + * are removed later by dce. So it is not a problem to just cleanup their derefs. */ + if (type->type == HLSL_CLASS_STRUCT || type->type == HLSL_CLASS_ARRAY) + { + hlsl_cleanup_deref(deref); + return; + } + if (!(offset = new_offset_instr_from_deref(ctx, &block, deref, &instr->loc))) return; list_move_before(&instr->entry, &block.instrs);

1 0

Francisco Casas : vkd3d-shader/hlsl: Store the type's register size for each register set.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: d07247249abe59b7278064b6262bc623b53eba9a URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/d07247249abe59b7278064b6262bc… Author: Francisco Casas <fcasas(a)codeweavers.com> Date: Fri Oct 28 12:23:05 2022 -0300 vkd3d-shader/hlsl: Store the type's register size for each register set. --- libs/vkd3d-shader/hlsl.c | 106 +++++++++++++++++++++++++++++++-------- libs/vkd3d-shader/hlsl.h | 34 +++++++++---- libs/vkd3d-shader/hlsl_codegen.c | 43 +++++++++------- libs/vkd3d-shader/hlsl_sm1.c | 2 +- libs/vkd3d-shader/hlsl_sm4.c | 4 +- 5 files changed, 137 insertions(+), 52 deletions(-)

1 0

Francisco Casas : vkd3d-shader/hlsl: Obtain extern resources as a separate array in SM4.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: 5272c5f86aa2511bcf8bc4a2682589156c807e93 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/5272c5f86aa2511bcf8bc4a268258… Author: Francisco Casas <fcasas(a)codeweavers.com> Date: Wed Feb 15 20:08:44 2023 -0300 vkd3d-shader/hlsl: Obtain extern resources as a separate array in SM4. --- libs/vkd3d-shader/hlsl_sm4.c | 92 +++++++++++++++++++++----------------------- 1 file changed, 43 insertions(+), 49 deletions(-) diff --git a/libs/vkd3d-shader/hlsl_sm4.c b/libs/vkd3d-shader/hlsl_sm4.c index 87882fc4..b7eb5ac9 100644 --- a/libs/vkd3d-shader/hlsl_sm4.c +++ b/libs/vkd3d-shader/hlsl_sm4.c @@ -480,52 +480,50 @@ static D3D_SRV_DIMENSION sm4_rdef_resource_dimension(const struct hlsl_type *typ } } -static int sm4_compare_externs(const struct hlsl_ir_var *a, const struct hlsl_ir_var *b) +static int sm4_compare_extern_resources(const void *a, const void *b) { - if (a->data_type->base_type != b->data_type->base_type) - return a->data_type->base_type - b->data_type->base_type; - if (a->reg.allocated && b->reg.allocated) - return a->reg.id - b->reg.id; - return strcmp(a->name, b->name); + const struct hlsl_ir_var *aa = *(const struct hlsl_ir_var **)a; + const struct hlsl_ir_var *bb = *(const struct hlsl_ir_var **)b; + + if (aa->data_type->base_type != bb->data_type->base_type) + return aa->data_type->base_type - bb->data_type->base_type; + return aa->reg.id - bb->reg.id; } -static void sm4_sort_extern(struct list *sorted, struct hlsl_ir_var *to_sort) +static const struct hlsl_ir_var **sm4_get_extern_resources(struct hlsl_ctx *ctx, unsigned int *count) { - struct hlsl_ir_var *var; - - list_remove(&to_sort->extern_entry); + const struct hlsl_ir_var **extern_resources = NULL; + const struct hlsl_ir_var *var; + size_t capacity = 0; + *count = 0; - LIST_FOR_EACH_ENTRY(var, sorted, struct hlsl_ir_var, extern_entry) + LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) { - if (sm4_compare_externs(to_sort, var) < 0) + if (var->reg.allocated && var->data_type->type == HLSL_CLASS_OBJECT) { - list_add_before(&var->extern_entry, &to_sort->extern_entry); - return; + if (!(hlsl_array_reserve(ctx, (void **)&extern_resources, &capacity, *count + 1, + sizeof(*extern_resources)))) + { + *count = 0; + return NULL; + } + + extern_resources[*count] = var; + ++*count; } } - list_add_tail(sorted, &to_sort->extern_entry); -} - -static void sm4_sort_externs(struct hlsl_ctx *ctx) -{ - struct list sorted = LIST_INIT(sorted); - struct hlsl_ir_var *var, *next; - - LIST_FOR_EACH_ENTRY_SAFE(var, next, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) - { - if (var->data_type->type == HLSL_CLASS_OBJECT) - sm4_sort_extern(&sorted, var); - } - list_move_tail(&ctx->extern_vars, &sorted); + qsort(extern_resources, *count, sizeof(*extern_resources), sm4_compare_extern_resources); + return extern_resources; } static void write_sm4_rdef(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc) { + unsigned int cbuffer_count = 0, resource_count = 0, extern_resources_count, i, j; size_t cbuffers_offset, resources_offset, creator_offset, string_offset; size_t cbuffer_position, resource_position, creator_position; - unsigned int cbuffer_count = 0, resource_count = 0, i, j; const struct hlsl_profile_info *profile = ctx->profile; + const struct hlsl_ir_var **extern_resources; struct vkd3d_bytecode_buffer buffer = {0}; const struct hlsl_buffer *cbuffer; const struct hlsl_ir_var *var; @@ -540,14 +538,9 @@ static void write_sm4_rdef(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc) 0x4353, /* COMPUTE */ }; - sm4_sort_externs(ctx); - - LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) - { - if (var->reg.allocated && var->data_type->type == HLSL_CLASS_OBJECT) - ++resource_count; - } + extern_resources = sm4_get_extern_resources(ctx, &extern_resources_count); + resource_count += extern_resources_count; LIST_FOR_EACH_ENTRY(cbuffer, &ctx->buffers, struct hlsl_buffer, entry) { if (cbuffer->reg.allocated) @@ -583,12 +576,11 @@ static void write_sm4_rdef(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc) resources_offset = bytecode_get_size(&buffer); set_u32(&buffer, resource_position, resources_offset); - LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) + for (i = 0; i < extern_resources_count; ++i) { uint32_t flags = 0; - if (!var->reg.allocated || var->data_type->type != HLSL_CLASS_OBJECT) - continue; + var = extern_resources[i]; if (var->reg_reservation.type) flags |= D3D_SIF_USERPACKED; @@ -633,12 +625,9 @@ static void write_sm4_rdef(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc) put_u32(&buffer, flags); /* flags */ } - i = 0; - - LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) + for (i = 0; i < extern_resources_count; ++i) { - if (!var->reg.allocated || var->data_type->type != HLSL_CLASS_OBJECT) - continue; + var = extern_resources[i]; string_offset = put_string(&buffer, var->name); set_u32(&buffer, resources_offset + i++ * 8 * sizeof(uint32_t), string_offset); @@ -653,8 +642,6 @@ static void write_sm4_rdef(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc) set_u32(&buffer, resources_offset + i++ * 8 * sizeof(uint32_t), string_offset); } - assert(i == resource_count); - /* Buffers. */ cbuffers_offset = bytecode_get_size(&buffer); @@ -747,6 +734,8 @@ static void write_sm4_rdef(struct hlsl_ctx *ctx, struct dxbc_writer *dxbc) set_u32(&buffer, creator_position, creator_offset); dxbc_writer_add_section(dxbc, TAG_RDEF, buffer.data, buffer.size); + + vkd3d_free(extern_resources); } static enum vkd3d_sm4_resource_type sm4_resource_dimension(const struct hlsl_type *type) @@ -2438,7 +2427,9 @@ static void write_sm4_shdr(struct hlsl_ctx *ctx, const struct hlsl_ir_function_decl *entry_func, struct dxbc_writer *dxbc) { const struct hlsl_profile_info *profile = ctx->profile; + const struct hlsl_ir_var **extern_resources; struct vkd3d_bytecode_buffer buffer = {0}; + unsigned int extern_resources_count, i; const struct hlsl_buffer *cbuffer; const struct hlsl_ir_var *var; size_t token_count_position; @@ -2456,6 +2447,8 @@ static void write_sm4_shdr(struct hlsl_ctx *ctx, VKD3D_SM4_LIB, }; + extern_resources = sm4_get_extern_resources(ctx, &extern_resources_count); + put_u32(&buffer, vkd3d_make_u32((profile->major_version << 4) | profile->minor_version, shader_types[profile->type])); token_count_position = put_u32(&buffer, 0); @@ -2465,10 +2458,9 @@ static void write_sm4_shdr(struct hlsl_ctx *ctx, write_sm4_dcl_constant_buffer(&buffer, cbuffer); } - LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, const struct hlsl_ir_var, extern_entry) + for (i = 0; i < extern_resources_count; ++i) { - if (!var->reg.allocated || var->data_type->type != HLSL_CLASS_OBJECT) - continue; + var = extern_resources[i]; if (var->data_type->base_type == HLSL_TYPE_SAMPLER) write_sm4_dcl_sampler(&buffer, var); @@ -2495,6 +2487,8 @@ static void write_sm4_shdr(struct hlsl_ctx *ctx, set_u32(&buffer, token_count_position, bytecode_get_size(&buffer) / sizeof(uint32_t)); dxbc_writer_add_section(dxbc, TAG_SHDR, buffer.data, buffer.size); + + vkd3d_free(extern_resources); } int hlsl_sm4_write(struct hlsl_ctx *ctx, struct hlsl_ir_function_decl *entry_func, struct vkd3d_shader_code *out)

1 0

Francisco Casas : vkd3d-shader/hlsl: Keep an hlsl_reg for each register set in hlsl_ir_var.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: e0031d2a1f40792ac85619a495bf5197f248b0e1 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/e0031d2a1f40792ac85619a495bf5… Author: Francisco Casas <fcasas(a)codeweavers.com> Date: Thu Nov 24 17:03:54 2022 -0300 vkd3d-shader/hlsl: Keep an hlsl_reg for each register set in hlsl_ir_var. --- libs/vkd3d-shader/hlsl.h | 16 ++++------- libs/vkd3d-shader/hlsl_codegen.c | 47 ++++++++++++++++++------------- libs/vkd3d-shader/hlsl_sm1.c | 22 ++++++++++----- libs/vkd3d-shader/hlsl_sm4.c | 60 ++++++++++++++++++++++++---------------- 4 files changed, 84 insertions(+), 61 deletions(-)

1 0

Francisco Casas : vkd3d-shader/hlsl: Allocate objects according to register set.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: c5384b38ee07015fdfcffbea7aa2ad3e1eb3a8ea URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/c5384b38ee07015fdfcffbea7aa2a… Author: Francisco Casas <fcasas(a)codeweavers.com> Date: Mon Jan 30 18:27:32 2023 -0300 vkd3d-shader/hlsl: Allocate objects according to register set. --- libs/vkd3d-shader/hlsl_codegen.c | 85 +++++++++++++++++----------------------- 1 file changed, 37 insertions(+), 48 deletions(-) diff --git a/libs/vkd3d-shader/hlsl_codegen.c b/libs/vkd3d-shader/hlsl_codegen.c index fd189b2a..d5534d8c 100644 --- a/libs/vkd3d-shader/hlsl_codegen.c +++ b/libs/vkd3d-shader/hlsl_codegen.c @@ -2234,6 +2234,22 @@ static void dump_function(struct rb_entry *entry, void *context) rb_for_each_entry(&func->overloads, dump_function_decl, ctx); } +static char get_regset_name(enum hlsl_regset regset) +{ + switch (regset) + { + case HLSL_REGSET_SAMPLERS: + return 's'; + case HLSL_REGSET_TEXTURES: + return 't'; + case HLSL_REGSET_UAVS: + return 'u'; + case HLSL_REGSET_NUMERIC: + vkd3d_unreachable(); + } + vkd3d_unreachable(); +} + /* Compute the earliest and latest liveness for each variable. In the case that * a variable is accessed inside of a loop, we promote its liveness to extend * to at least the range of the entire loop. Note that we don't need to do this @@ -2885,50 +2901,28 @@ static void allocate_buffers(struct hlsl_ctx *ctx) } } -static const struct hlsl_ir_var *get_reserved_object(struct hlsl_ctx *ctx, char type, uint32_t index) +static const struct hlsl_ir_var *get_reserved_object(struct hlsl_ctx *ctx, enum hlsl_regset regset, + uint32_t index) { const struct hlsl_ir_var *var; LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, const struct hlsl_ir_var, extern_entry) { - if (var->last_read && var->reg_reservation.type == type && var->reg_reservation.index == index) + if (var->last_read && var->reg_reservation.type == get_regset_name(regset) + && var->reg_reservation.index == index) return var; } return NULL; } -static const struct object_type_info -{ - enum hlsl_base_type type; - char reg_name; -} -object_types[] = -{ - { HLSL_TYPE_SAMPLER, 's' }, - { HLSL_TYPE_TEXTURE, 't' }, - { HLSL_TYPE_UAV, 'u' }, -}; - -static const struct object_type_info *get_object_type_info(enum hlsl_base_type type) -{ - unsigned int i; - - for (i = 0; i < ARRAY_SIZE(object_types); ++i) - if (type == object_types[i].type) - return &object_types[i]; - - WARN("No type info for object type %u.\n", type); - return NULL; -} - -static void allocate_objects(struct hlsl_ctx *ctx, enum hlsl_base_type type) +static void allocate_objects(struct hlsl_ctx *ctx, enum hlsl_regset regset) { - const struct object_type_info *type_info = get_object_type_info(type); + char regset_name = get_regset_name(regset); struct hlsl_ir_var *var; uint32_t min_index = 0; uint32_t index; - if (type == HLSL_TYPE_UAV) + if (regset == HLSL_REGSET_UAVS) { LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) { @@ -2942,21 +2936,17 @@ static void allocate_objects(struct hlsl_ctx *ctx, enum hlsl_base_type type) LIST_FOR_EACH_ENTRY(var, &ctx->extern_vars, struct hlsl_ir_var, extern_entry) { - enum hlsl_regset regset; - - if (!var->last_read || var->data_type->type != HLSL_CLASS_OBJECT - || var->data_type->base_type != type) + if (!var->last_read || var->data_type->reg_size[regset] == 0) continue; - regset = hlsl_type_get_regset(var->data_type); - - if (var->reg_reservation.type == type_info->reg_name) + if (var->reg_reservation.type == regset_name) { - const struct hlsl_ir_var *reserved_object = get_reserved_object(ctx, type_info->reg_name, + const struct hlsl_ir_var *reserved_object = get_reserved_object(ctx, regset, var->reg_reservation.index); if (var->reg_reservation.index < min_index) { + assert(regset == HLSL_REGSET_UAVS); hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_OVERLAPPING_RESERVATIONS, "UAV index (%u) must be higher than the maximum render target index (%u).", var->reg_reservation.index, min_index - 1); @@ -2964,25 +2954,24 @@ static void allocate_objects(struct hlsl_ctx *ctx, enum hlsl_base_type type) else if (reserved_object && reserved_object != var) { hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_OVERLAPPING_RESERVATIONS, - "Multiple objects bound to %c%u.", type_info->reg_name, - var->reg_reservation.index); + "Multiple objects bound to %c%u.", regset_name, var->reg_reservation.index); hlsl_note(ctx, &reserved_object->loc, VKD3D_SHADER_LOG_ERROR, - "Object '%s' is already bound to %c%u.", reserved_object->name, - type_info->reg_name, var->reg_reservation.index); + "Object '%s' is already bound to %c%u.", reserved_object->name, regset_name, + var->reg_reservation.index); } var->regs[regset].id = var->reg_reservation.index; var->regs[regset].allocated = true; - TRACE("Allocated reserved %s to %c%u.\n", var->name, type_info->reg_name, var->reg_reservation.index); + TRACE("Allocated reserved %s to %c%u.\n", var->name, regset_name, var->regs[regset].id); } else if (!var->reg_reservation.type) { - while (get_reserved_object(ctx, type_info->reg_name, index)) + while (get_reserved_object(ctx, regset, index)) ++index; var->regs[regset].id = index; var->regs[regset].allocated = true; - TRACE("Allocated object to %c%u.\n", type_info->reg_name, index); + TRACE("Allocated object to %c%u.\n", regset_name, index); ++index; } else @@ -2992,7 +2981,7 @@ static void allocate_objects(struct hlsl_ctx *ctx, enum hlsl_base_type type) type_string = hlsl_type_to_string(ctx, var->data_type); hlsl_error(ctx, &var->loc, VKD3D_SHADER_ERROR_HLSL_INVALID_RESERVATION, "Object of type '%s' must be bound to register type '%c'.", - type_string->buffer, type_info->reg_name); + type_string->buffer, regset_name); hlsl_release_string_buffer(ctx, type_string); } } @@ -3318,11 +3307,11 @@ int hlsl_emit_bytecode(struct hlsl_ctx *ctx, struct hlsl_ir_function_decl *entry else { allocate_buffers(ctx); - allocate_objects(ctx, HLSL_TYPE_TEXTURE); - allocate_objects(ctx, HLSL_TYPE_UAV); + allocate_objects(ctx, HLSL_REGSET_TEXTURES); + allocate_objects(ctx, HLSL_REGSET_UAVS); } allocate_semantic_registers(ctx); - allocate_objects(ctx, HLSL_TYPE_SAMPLER); + allocate_objects(ctx, HLSL_REGSET_SAMPLERS); if (ctx->result) return ctx->result;

1 0

Francisco Casas : tests: Test allocation of unused objects.
by Alexandre Julliard 22 Feb '23

22 Feb '23

Module: vkd3d Branch: master Commit: 7a7116eaab73b4becd9d526702aa9c1998dcda31 URL: https://gitlab.winehq.org/wine/vkd3d/-/commit/7a7116eaab73b4becd9d526702aa9… Author: Francisco Casas <fcasas(a)codeweavers.com> Date: Thu Feb 9 17:43:49 2023 -0300 tests: Test allocation of unused objects. --- Makefile.am | 1 + tests/register-reservations.shader_test | 51 +++++++++++++++++++++++++++++++++ tests/shader_runner_d3d12.c | 2 +- 3 files changed, 53 insertions(+), 1 deletion(-) diff --git a/Makefile.am b/Makefile.am index c8812892..ebda192d 100644 --- a/Makefile.am +++ b/Makefile.am @@ -135,6 +135,7 @@ vkd3d_shader_tests = \ tests/preproc-macro.shader_test \ tests/preproc-misc.shader_test \ tests/reflect.shader_test \ + tests/register-reservations.shader_test \ tests/return.shader_test \ tests/round.shader_test \ tests/sampler.shader_test \ diff --git a/tests/register-reservations.shader_test b/tests/register-reservations.shader_test new file mode 100644 index 00000000..04de717b --- /dev/null +++ b/tests/register-reservations.shader_test @@ -0,0 +1,51 @@ +[require] +shader model >= 4.0 + + +[texture 0] +size (1, 1) +0.0 0.0 0.0 99.0 + +[texture 1] +size (1, 1) +1.0 1.0 1.0 99.0 + +[texture 2] +size (1, 1) +2.0 2.0 2.0 99.0 + +[texture 3] +size (1, 1) +3.0 3.0 3.0 99.0 + +[texture 4] +size (1, 1) +4.0 4.0 4.0 99.0 + + +[pixel shader] +Texture2D unused : register(t0); +Texture2D tex; + +float4 main() : sv_target +{ + return tex.Load(int3(0, 0, 0)); +} + +[test] +draw quad +todo probe all rgba (1.0, 1.0, 1.0, 99.0) + + +[pixel shader] +Texture2D unused[2][2] : register(t0); +Texture2D tex; + +float4 main() : sv_target +{ + return tex.Load(int3(0, 0, 0)); +} + +[test] +draw quad +todo probe all rgba (4.0, 4.0, 4.0, 99.0) diff --git a/tests/shader_runner_d3d12.c b/tests/shader_runner_d3d12.c index bd94b4c9..3e661151 100644 --- a/tests/shader_runner_d3d12.c +++ b/tests/shader_runner_d3d12.c @@ -167,7 +167,7 @@ static ID3D12RootSignature *d3d12_runner_create_root_signature(struct d3d12_shad ID3D12GraphicsCommandList *command_list, unsigned int *uniform_index) { D3D12_ROOT_SIGNATURE_DESC root_signature_desc = {0}; - D3D12_ROOT_PARAMETER root_params[4], *root_param; + D3D12_ROOT_PARAMETER root_params[5], *root_param; D3D12_STATIC_SAMPLER_DESC static_samplers[1]; ID3D12RootSignature *root_signature; HRESULT hr;

1 0

Alexandros Frantzis : gdi32: Track ticks since draw start per window_surface.
by Alexandre Julliard 21 Feb '23

21 Feb '23

Module: wine Branch: master Commit: aa72d7d42d64a8fd9ded6d870f261eeaccf632fb URL: https://gitlab.winehq.org/wine/wine/-/commit/aa72d7d42d64a8fd9ded6d870f261e… Author: Alexandros Frantzis <alexandros.frantzis(a)collabora.com> Date: Thu Feb 16 18:17:03 2023 +0200 gdi32: Track ticks since draw start per window_surface. Track ticks since draw start per window_surface, instead of per DC as is currently the case. This change helps reduce visual glitches caused by badly timed flushes when drawing to the same window_surface from multiple DCs (e.g., for child windows). This approach is a better fit for the current heuristic for forcing flushing, which consults the shared window_surface bounds to decide whether this is the start of a draw in order to reset the (currently per DC) draw start time. The problem in the current implementation occurs when a drawing to a DC begins with an already damaged window_surface, e.g., due to draws from other DCs targeting that window_surface. In such a case, the DC draw start time is not reset and refers to the start of some previous draw sequence using this DC, thus increasing the chances that the 50ms time flush limit will be eventually exceeded in the middle of the current draw sequence. In other words, the state of the (shared) window_surface damage is not a reliable indicator of the beginning (or not) of a draw to a DC. An example, assuming DC1 and DC2 target the same window_surface: DC1.start_ticks = 0 DC2.start_ticks = 0 FLUSH_PERIOD = 50 0 flush 1 draw to DC1 -> DC1.start_ticks = 1 ... [no flush] ... 2 draw to DC2 -> DC2.start_ticks remains 0 since surface is damaged ... 50 flush 51 draw to DC1 -> DC1.start_ticks = 51 ... [no flush] ... 52 draw to DC2 -> DC2.start_ticks remains 0 since surface is damaged, current - DC2.start_ticks > FLUSH_PERIOD so we are forced to flush in the middle of the drawing sequence => potential glitch Tracking the draw start per window_surface ameliorates the problem because the beginning of a draw on a DC targeting an undamaged window_surface resets the start time for all DCs targeting that window_surface: ... 50 flush 51 draw to DC1 -> surface.draw_ticks = 51 ... [no flush] ... 52 draw to DC2 -> surface.draw_ticks remains 51 since surface is damaged, but current - surface.draw_ticks < FLUSH_PERIOD, so we do not flush Signed-off-by: Alexandros Frantzis <alexandros.frantzis(a)collabora.com> --- dlls/win32u/dibdrv/dc.c | 6 +++--- include/wine/gdi_driver.h | 1 + 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/dlls/win32u/dibdrv/dc.c b/dlls/win32u/dibdrv/dc.c index 68a009f576e..f96374c1229 100644 --- a/dlls/win32u/dibdrv/dc.c +++ b/dlls/win32u/dibdrv/dc.c @@ -731,7 +731,6 @@ struct windrv_physdev struct gdi_physdev dev; struct dibdrv_physdev *dibdrv; struct window_surface *surface; - DWORD start_ticks; }; static const struct gdi_dc_funcs window_driver; @@ -745,13 +744,14 @@ static inline void lock_surface( struct windrv_physdev *dev ) { /* gdi_lock should not be locked */ dev->surface->funcs->lock( dev->surface ); - if (IsRectEmpty( dev->dibdrv->bounds )) dev->start_ticks = NtGetTickCount(); + if (IsRectEmpty( dev->dibdrv->bounds )) dev->surface->draw_start_ticks = NtGetTickCount(); } static inline void unlock_surface( struct windrv_physdev *dev ) { + BOOL should_flush = NtGetTickCount() - dev->surface->draw_start_ticks > FLUSH_PERIOD; dev->surface->funcs->unlock( dev->surface ); - if (NtGetTickCount() - dev->start_ticks > FLUSH_PERIOD) dev->surface->funcs->flush( dev->surface ); + if (should_flush) dev->surface->funcs->flush( dev->surface ); } static void CDECL unlock_bits_surface( struct gdi_image_bits *bits ) diff --git a/include/wine/gdi_driver.h b/include/wine/gdi_driver.h index 3de7e20af9a..fe0894b2820 100644 --- a/include/wine/gdi_driver.h +++ b/include/wine/gdi_driver.h @@ -217,6 +217,7 @@ struct window_surface struct list entry; /* entry in global list managed by user32 */ LONG ref; /* reference count */ RECT rect; /* constant, no locking needed */ + DWORD draw_start_ticks; /* start ticks of fresh draw */ /* driver-specific fields here */ };

1 0

Alexandros Frantzis : win32u: Reset draw_start_ticks for new window_surface.
by Alexandre Julliard 21 Feb '23

21 Feb '23

Module: wine Branch: master Commit: 6d2bd75c4a82d10ee98c63010a5701d428ccc5d0 URL: https://gitlab.winehq.org/wine/wine/-/commit/6d2bd75c4a82d10ee98c63010a5701… Author: Alexandros Frantzis <alexandros.frantzis(a)collabora.com> Date: Thu Feb 16 18:20:56 2023 +0200 win32u: Reset draw_start_ticks for new window_surface. Normally, a new window_surface has no damage so its draw_start_ticks value is reset at the time of the first lock_surface. However, if contents from the old window_surface are copied to the new one when the window_surface for a window is updated, the new window_surface will be initially damaged. We want to reset the draw_start_ticks value in this case too, to avoid flushes occuring soon after such updates, during the initial redraw of the window_surface. Signed-off-by: Alexandros Frantzis <alexandros.frantzis(a)collabora.com> --- dlls/win32u/dibdrv/dc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/dlls/win32u/dibdrv/dc.c b/dlls/win32u/dibdrv/dc.c index f96374c1229..23ef7f64390 100644 --- a/dlls/win32u/dibdrv/dc.c +++ b/dlls/win32u/dibdrv/dc.c @@ -744,7 +744,8 @@ static inline void lock_surface( struct windrv_physdev *dev ) { /* gdi_lock should not be locked */ dev->surface->funcs->lock( dev->surface ); - if (IsRectEmpty( dev->dibdrv->bounds )) dev->surface->draw_start_ticks = NtGetTickCount(); + if (IsRectEmpty( dev->dibdrv->bounds ) || dev->surface->draw_start_ticks == 0) + dev->surface->draw_start_ticks = NtGetTickCount(); } static inline void unlock_surface( struct windrv_physdev *dev )

1 0